This patent by Digimarc Corporation gives Brainchip Akida a mention:
Worth noting that it mentions a conveyor system, which was mentioned by Brainchip in the past but hasn't been mentioned again for some time.
Also interesting that someone was mentioning just the other day how they'd like an Akida system to identify plastics or the like which are suitable for recycling. This is a solution for a larger recycling factory. Sounds like beneficial AI to me.
https://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.html&r=1&f=G&l=50&d=PG01&p=1&S1=(((%22Van+Der+Made%22.IN.)+AND+%22Peter%22)+OR+BrainChip)&OS=IN/(%22Van+Der+Made%22)+AND+%22Peter%22+OR+BrainChip&RS=((IN/%22Van+Der+Made%22+AND+%22Peter%22)+OR+BrainChip)
| Inventors: | Filler; Tomas;(Beaverton, OR); Holub; Vojtech;(Portland, OR); Sharma; Ravi K.;(Portland, OR); Rodriguez; Tony F.;(Portland, OR); Alattar; Osama M.;(Tigard, OR); Alattar; Adnan M.;(Tigard, OR); Lord; John D.;(West Linn, OR); Johnson; Brian;(Vancouver, WA); Ruotolo; David;(Portland, OR); Rhoads; Geoffrey B.;(West Linn, OR); Brunk; Hugh L.;(Portland, OR); Sedighianaraki; Vahid;(Portland, OR) |
|---|
| Applicant: | | Name | City | State | Country | Type |
|---|
Digimarc Corporation |
Beaverton |
OR |
US |
|
|
|---|
| Family ID: | 1000005638298 |
| Appl. No.: | 17/214455 |
| Filed: | March 26, 2021 |
| United States Patent Application | 20210299706 |
| Kind Code | A1 |
| Filler; Tomas ; et al. | September 30, 2021 |
ARRANGEMENTS FOR DIGITAL MARKING AND READING OF ITEMS, USEFUL IN RECYCLING
AbstractImages depicting items in a waste flow on a conveyor belt are provided to two analysis systems. The first system processes images to decode digital watermark payload data found on certain of the items (e.g., plastic containers). This payload data is used to look up corresponding attribute metadata for the items in a database, such as the type of plastic in each item, and whether the item was used as a food container or not. The second analysis system can be a spectroscopy system that determines the type of plastic in each item by its absorption characteristics. When the two systems conflict in identifying the plastic type, a sorting logic processor applies a rule set to arbitrate the conflict and determine which plastic type is most likely. The item is then sorted into one of several different bins depending on a combination of the final plastic identification, and whether the item was used as a food container or not. A variety of other features and arrangements are also detailed.
Claims
1. A waste sorting apparatus comprising: a conveyor; one or more cameras, and one or more light sources, to produce imagery of items on the conveyor; a first identification system that applies watermark decoding to said imagery to obtain first information about an item on the conveyor; a second identification system that applies spectroscopy or neural network classification to said imagery to obtain second information about said item on the conveyor; one or more diverters; and a sorting logic processor coupled to both the first and second identification systems, configured to control the one or more diverters in accordance with output data provided by said first and second identification systems.
2. The apparatus of claim 1 in which the sorting logic processor is configured to respond to conflicting information provided by the first and second identification systems, by giving precedence to the first information in a first circumstance, and by giving precedence to the second information in a second circumstance.
3. The apparatus of claim 1 in which the sorting logic processor is configured to respond to conflicting information provided by the first and second identification systems, by giving precedence to the first information in a circumstance in which the item is black in color, and by giving precedence to the second information in a circumstance in which the item is not black in color.
4. A waste sorting apparatus comprising: a near infrared imaging system including one or more processors configured to discern a spectral signature produced by a plastic container, and to identify a plastic resin of the container based on said spectral signature; a watermark imaging system including one or more processors configured to extract an encoded watermark payload formed in a surface of said plastic container, or printed on a label of said plastic container, and to determine information from said watermark payload; and a processing system configured to make a sorting decision for said plastic container based on the plastic resin identified by the near infrared imaging system, and based on the information determined from the watermark payload.
5. The waste sorting apparatus of claim 4 in which the processing system is further configured to resolve a conflict between the plastic resin identified by the near infrared imaging system, and a plastic resin determined from said watermark payload.
6. The waste sorting apparatus of claim 4 in which the information determined from the watermark payload includes whether the container was used for food or not, and the processing system is configured to cause the container to be sorted based on both the identified plastic resin, and whether the container was used for food or not.
7. The waste sorting apparatus of claim 4 in which the information determined from the watermark payload includes whether the container was formed of virgin or non-virgin plastic, and the processing system is configured to cause the container to be sorted based on both the identified plastic resin, and whether the container was formed of virgin or non-virgin plastic.
8. A waste sorting apparatus comprising: an artificial intelligence system that processes data, including image data, to make a judgment about an item in a waste stream; a watermark system including a camera and one or more processors configured to extract an encoded watermark payload formed in a surface of said item, or printed on a label applied to said item, and to determine information from said watermark payload; and a diverter that sorts said item from the waste stream based on the judgment made by the artificial intelligence system and based on the information determined from said watermark payload.
CONCLUDING REMARKS
[0342] Still another type of processor hardware is a neural network chip, e.g., the Intel Nervana NNP-T, NNP-I and Loihi chips, the Google Edge TPU chip, and theBrainchipAkida neuromorphic SOC.
(20min delay)